Artificial Intelligence and Neural Networks
Copernicus, the European Union’s Earth observation programme, and the commercial Planet-Scope data together provide extensive and steadily growing data archives. Using them for earth observation requires appropriate analysis methods and sets new requirements for algorithms and computing capacity.
Deep learning, a subdomain of artificial intelligence, has proven to be an adequate concept for analyzing large volumes of data (big data). These data archives permit neural networks to learn and at the same time form the basis for extensive analyses and efficient monitoring. A particular highlight: The system can be adapted very easily to all sorts of different geographical conditions.
- Identification of road and housing infrastructure
- Classification and monitoring of land cover changes
- Monitoring of single objects such as ships, vehicles or aircrafts.
- Activity based intelligence to understand the “Patterns-of Live” for security-relevant zones such as airports, borders, ports.
- Early crisis warning through spatial and temporal correlation across multiple data sets